winstonp

Same. Overuse of the word "genuinely" is another tell.

Which apps have you seen ask for someone to setup a local LLM? Can't recall having ever seen one

The new Studio Displays have more powerful chips than the Neo.

Agree. Claude tends to produce better design, but from a system understanding and architecture perspective Codex is the far better model

Everybody says they are, with the main point being they can get the M6 Pros onto TSMC's 2nm node and save 3nm capacity for iPhones.

They are at least nice for comparing it with the max of the Intel. That should really say gives them up to 22 additional hours given the wear on their batteries lol

DeepSeek hasn't been SotA in at least 12 calendar months, which might as well be a decade in LLM years

This guy was into the bagscoin BS that was going on on X this past month. Wouldn't trust a word he says.

Once Chrome does, for many devs, you can simply enforce a version check and say "please use latest Chrome" and be done with it.

yep. i had a hell of a time building a scheduling system for a business that worked across timezones with the date API

the new Date() constructor is an amalgamation of like 5 different specs, and unless the input matches one of them, which one kicks in is up to the implementer's choice

won't? i doubt that. can't tell you the last time i paid $20/mo for any non-business sub, but i've paid for some combo of openai+claude for last year

I also built a RAG system about a year back for technical search, everything seems the same!

Rumor has it M6 Pro will be a total redesign. Whether that's a good or bad thing depends on how much you trust Apple to nail a next gen design first try again

when I did startups, we had multiple companies who would not sign deals until our SOC2 was complete

AFAIK the 757 frame is too heavy to be powered by the LEAP engines. Those planes were powered by a class of engine between the old 737 and 777 engines, and nobody makes them anymore because they're not in demand, so a 757 MAX is just not financially viable.

It's called TanStack Start

i am just simply not touching tailwind v3 -> 4. v4 is for new projects only.

Mine is, on iOS:

* in safari private mode, open image picker

* switch to different app (e.g. go to WhatsApp to save a new image)

* go back to safari

the image picker can now no longer be spawned from that safari private tab, you'll have to open a new tab to re enable the image picker.

I agree Go error handling is unoptimal, but this is simply not the right approach. This essentially turns error handling into a whole other language, almost like how Ginkgo is a separate language for handling tests.

OpenAI has been down, but Claude and Meta services are working fine for me

Been using it, though it having a 150 swipe limit is pretty funny for a local app that doesn't use any cloud or compute resources.

It's not. Your problems with Face ID seem to be of a super minority of users -- 99% of people I know have no problems with it.

Would love one that looks as slick as Supabase's web ui

Curious to see if ignoring robots.txt, which has almost miraculously held up for decades, could eventually lead to a lawsuit.

I do cmd+ctrl+v

It is incredible how poor the caps lock experience is on Linux in 2024. What setup do you use for that?